ASSAY FOR 1,25-DIHYDROXY VITAMIN D RECEPTOR PROTEIN This invention was made with United States govern¬ ment support awarded by the National Institute of Health (NIH), Grant number DK-14881. The United States Govern- ment has certain rights in this invention.

Technical Field

This invention relates to an assay for testing for the level of 1,25-dihydroxy vitamin D receptor protein (e.g. in mammalian samples). More specifically it involves the use of immunoassay techniques as a diagnos¬ tic tool.

Background Art

Vitamin D is a well known vitamin which has many useful functions in mammals. It is activated by 25- hydroxylation in the liver and subsequently by 1-hydroxy- lation in the kidney. The active form of vitamin D, 1,25- dihydroxyvitamin D3 (1,25-(OH)2D3) regulates plasma calcium and phosphorus levels by acting on the intestine, bone and the kidneys. Other possible target organs include

endocrine giands, skin, reproductive organs and the hematopoietic system.

Actions of the l,25-(OH)2D3 hormone are believed to be mediated through an intracellular receptor protein to which it binds. See generally J. . Pike, 43 Nutr. Rev. 161-168 (f§85). The disclosure of this article and all other, articles recited herein are incorporated by refer¬ ence as if fully set forth herein. After binding of the hormone to the receptor protein, the receptor-hormone complex is thought to bind to specific DNA sequences and alter gene transcription.

Because of important known and proposed actions of vitamin D, it is of great interest to understand in a quantitative*way the regulation, distribution, and turnover of its receptor protein. There is also reason to believe that the level of receptor protein will provide information as to the existence or stage of a disease, or its treatment (e.g. some cancers), and/or provide a means for monitoring and selecting drug treatments based on vitamin D variants.

Existing prior art techniques for assaying for vitamin D receptor rely on binding of radioactive vitamin D or on immunoblotting techniques. See generally W. ecksler et al. , 92 Anal. Biochem. 314-323 (1979); R. Shepard et al. , 182 J. Biochem. 55-69 (1979); S. Dokoh et al. , 221 J. Biochem. 129-136 (1984); and J. Gershoni ^' et al. , 131 Anal. Biochem. 1-15 (1983).

-3-

In a typical radioactive vitamin D assay, radio- labelled 1,25 vitamin D (the ligand) is added to the sample being tested. After binding, the bound complex is separated from unbound labelled 1,25 vitamin D with a compound called hydroxylapatite, and the radioactivity present in the bound fraction (and thus the receptor level) is measured. However, this technique measures only non-denatured forms of the receptor (and thus usually undercountε in an unpredictable manner), and - ay also in some cases be disrupted by the actions of vitamin D transport protein present in the sample. Immunoblotting techniques use iodine labelled antibody. While they do measure denatured receptor, they provide only a crude approximation of vitamin D3 receptor content.

Immunological assays using antibodies have been successfully employed for certain other steroid hormone receptors such as estrogen and progesterone receptors. G. Leclecq et aj^, 46 Cancer Res. 4233-4236 (1986) and S. Brailly et al. , li6 J. Endocrinol. 427-434 (1988). See also L. Miles et al^, 219 Nature 186-189 (1968). Further, there is now available purified receptor protein (see T. Brown et a , 85 P.N.A.S. USA 2454-2458 (1988)) and monoclonal antibodies that bind to different epitopes on the receptor. M. Dame et al. , 25 Biochem. 4523-34

(1986). However, to date no satisfactory radioimmunoas- say has been developed for receptor protein that is

sensitive, reproducible, easy to use, and useful in connection with crude samples from mammalian sources.

Disclosure Of Invention

One aspect of the invention provides a binding assay to determine the amount of 1,25-dihydroxy vitamin D receptor in a sample. One adds to the sample a plurality of a first labelled antibody that is capable of binding at a first epitope on the receptor and a plurality of second antibody that is capable of binding at a second epitope on the receptor. Using means connectable to the second antibody, one then separates at least a portion of receptor protein that is bound to the labelled antibody from unbound labelled antibody. Thereafter, one measures the amount of labelled antibody present in one of the separated fractions.

In a preferred form, the separation step is performed using immunoprecipitation (e.g. a biotin/avidin sepharose system), the label is a radioactive label, and the fraction that is measured is the receptor bound fraction. The sample is preferably a human sample, but it may also come from animal sources or be part of a control experiment.

Kits for conducting such assays are also provided. They have a first radioactively labelled antibody capable of binding to a first epitope of the receptor protein and a second antibody capable of binding to a second epitope

-5-

of the receptor protein. The second antibody can be biotinylated to render it readily immunoprecipitatable, and the first antibody can be radiolabelled with radio¬ active iodine (or other radioactive labels). In a preferred form, the kit also has 1,25-dihydroxy vitamin D receptor protein (to prepare a standard curve).

Thus, the invention focuses on the use of one labelled antibody and one anchorable antibody. They bind at different places on receptor protein, substantially without regard to denaturing. The "anchorable" antibody forms a means for the complex to be easily precipitated away from excess unbound labelled antibody (thus facili¬ tating measurement) .

It will be appreciated that in developing the present invention, the inventors had to overcome several problems. For one thing, vitamin D receptor protein is very unstable and will denature under many conditions. Also, tumor cell tissue will have varying degrees of vitamin D present, which will to varying degrees occupy the receptor. Further, techniques for anchoring the antibody in a way that produces reproducible results had to be developed. Other problems also had to be faced and overcome.

The objects of the invention therefore include: (a) providing an assay of the above kind that is sensitive, reproducible, and easy to use;

(b) providing kits that are of assistance in performing assays of the above kind.

Still other objects and advantages of the present inven¬ tion will be apparent from the description which follows.

Brief Description Of The Drawing

A better understanding of the present invention will be accomplished by reference to the drawing. It should be understood, however, that the drawing and the descrip¬ tion of the preferred embodiments which follow it are merely examples of the invention. They are not intended to represent the full scope of the invention. Rather, the claims should be looked to in order to determine the full scope of the invention.

Fig. 1 is a schematic depiction of a portion of a preferred embodiment of the assay.

Best Modes For Carrying Out The Invention

A. Buffers

Phosphate-buffered saline (PBS), 1.5 mM KH2PO4/8.I mM Na2HPθ4 (pH 8.0)/137 mM KC1; PBS-Triton, 0.5% Triton X-100 in PBS; TE, 50 mM Tris-HCl

(pH 7.4J/1.5 mM EDTA; TED, TE/5 mM dithiothreitol; TEDK 300, TED/300 mM KC1; TEDK 300-BSA, TEDK 300/0.5% (wt/vol) BSA/0.02% NaN3-

-7-

B . Monoclonal Antibodies

A first antibody (1) and a second antibody (7), IVG8C11 and VD2F12, were generated to the porcine intestinal l,25-(OH)2D3 receptor (3) using the techniques described in M. Dame et al. , 25 Biochem. 4523-34 (1986). Hybridomas capable of producing such antibodies can then be conveniently prepared as described by Dame et al■ , 25 Biochemistry 4523-34 (1986). Furthermore, a method for determining antibodies that bind to different epitopes (4) (5) is provided in this article. The antibodies produced by the hybridomas can be produced in mouse ascites and purified by chromatography on goat anti-mouse IgG-Sepharσse (HyClone, Logan, UT). Both antibodies cross-react with receptor from a variety of species, including human. Purified antibody can be concentrated on an Amicon Ultrafiltration Cell and stored at -20°C.

C. Biotinylation Of Antibody VD2F12 ^'

Purified VD2F12 (7) was coupled with biotin (6) using N-hydroxysuccinimidobiotin (Sigma, St. Louis, MO), 0.1 M in N,N-dimethylformamide (Aldrich, Milwaukee, WI), 10 μl/mg antibody, for 1 hour at room temperature followed by extensive dialysis with phosphate buffered saline (PBS). For general biotinylation techniques, see

D. Hullet, Ph.D. Thesis, University Of Wisconsin-Madison 180-204 (1984).

D. Iodination Of Antibody IVG8C11.

Purified IVG8C11 (1), 5 μg in 10 μl PBS, was reacted with 1.0 mCi ¹²⁵ I (2) Bolton-Hunter Reagent (DuPont/NEN, Wilmington, DE) for 2 hours on ice. See A. Bolton et al. , 133 J. Biochem. 529-533 (1973) for general iodina¬ tion techniques. The reaction was stopped with 0.25 ml 0.2 M glycine in 0.1 M borate (pH 8.5), for 5 minutes followed by addition of 0.25 ml PBS + 0.25% (wt/vol) gelatin. The mixture was then passed over a G-25 Sephadex column (Pharmacia, Uppsala, Sweden) saturated with BSA and pre-equilibrated with PBS + 0.25% gelatin, to separate antibody bound from free ^•J - ^ΛJ I. Antibody preparations had a specific activity of 5-10 μCi/μg.

E. Preparation Of Avidin-Sepharose Avidin from egg white (Calbioche , San Diego, CA) was coupled to Sepharose CL 4B (Pharmacia, Uppsala, Sweden) at a concentration of 0.5 mg avidin per ml wet Sepharose, using the general cyanogenbromide activation procedure of Kohn et al. , 107 Biochem. Biophys. Res. Coiramm. 878-884 (1982).

F. Preparation Of Pig Nuclear Exact (PNE)

To create one mammalian test sample, a high salt, crude nuclear extract from pig intestinal mucosa was obtained as previously described. See M. Dame et al. , 82 P.N.A.S. USA 7825-29 (1985).

G. Cytosol Preparation

Weanling rats were obtained from the Holtzman Co. (Madison, WI) and kept on a diet containing 0.47% calcium, 0.30% phosphorus and no vitamin D. Intestinal whole cell extracts were prepared from 12-week-old rats as previously described. E. Pierce et al. , 261 Arch. Biochem. Biophys. 241-49 (1988).

Biopsies of human duodenal mucosa from normal female volunteers (premenopausal) were obtained from the Mayo Clinic (Rochester, MN) . Intestinal mucosal extracts were prepared using the same protocol as for rat intestinal mucosa. See generally A. Bolton e_t al. , 133 J. Biochem. 529-39 (1973).

In the alternative, human tissue samples can be minced with a razor blade on ice, and washed by centrifu- gation (800 x g) 4 times with TENa 150 + 1 mM DTT. After weighing, the tissue can be homogenized on ice by a glass-teflon system or polytron (depending on source) in 2 volumes of TEK300 + 5 mM DTT + 5 mM DFP. They are then ultracentrifuged 1 hour at 170,000 x g. Supernatants are saved at -70°C until use. H. Preparation Of Purified l,25-(OH)?D^ Receptor

Pig nuclear extract was subjected to immunoaffinity chromatography (T. Brown et al. , 85 P.N.A.S. USA 2454-58 (1988)) and sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis (U. Laemmli, 227 Nature 680-85 (1970)), and electroelution. The purified receptor

protein concentration was then determined by soft laser scanning densitometry (model SL-504-XL) (Biomed Instru¬ ments, Fullerton, CA) of a single 55 kDa band on Coomassie Blue stained SDS-polyacryal ide gels using ovalbumin, bovine serum albumin and carbonic anhydrase as standards-(Sigma, St. Louis, MO). Purified receptor (3) was stored in aliquots at -70°C. I. Protein Determination

Protein was determined by the technique of M. Bradford, 72 Anal. Biochem. 248-254 (1976) using bovine serum albumin as a standard. J. Vitamin D Compounds

Radioactive l,25-(OH)2-[26,27- ³ H]D3 (160 Ci/mmol, 1 Ci=37 GBq) (for purposes of running a prior art assay for comparison) was produced by DuPont/NEN and prepared as previously described. J. Napoli et al. , 19 Biochem¬ istry 2515-2521 (1980). Nonradioactive l,25-(OH)2D3 was a gift from the Hoffmann-La Roche Company.

The Assay

For our reagents we used biotinylated antibody at 20 μg/ml (25 μL = 0.5 μg per tube); iodinated antibody ~1.4 x ID ⁶ cpm/ml (25 μL = 35000 cpm per tube); and then the test or the standard sample.

The reseptor standard was prepared as follows (1500 μL of 0.5 fmol/μL = 750 fmol total). We made- standards in 2.5 fmol increments by taking 20/40/60 ....

200 μL of concentrated standard (0.5 fmol/μL) in a total of 400 μL. Test samples can be diluted as desired and preferably 100 μL is used per assay tube. It is recom¬ mended to start at lx with unknown samples, and dilute to 1:8 as required.

For example, biotinylated VD2F12 (0.5 μg) (6) (7) was incubated with iodinated IVG8C11 (40,000 cpm) (1)(2) and purified receptor standard (0-25 fmol) (3) (or alterna¬ tively a sample) for 16 hours at +4°C in Beckman minitubes (Beckman, Palo Alto, CA) . Assay buffer for this example was TEDK 300 + 0.5% BSA and incubation volume 150 μl. Next a 50% slurry of avidin-Sepharose (8) in TE was added and incubated for 1 hour on ice (100 μl) with mixing every 20 minutes. The tubes were centrifuged at 800 x g for 5 minutes. The resulting pellets were washed three times with PBS-Triton and counted in Packard Multi-Prias Auto-Gamma counter (Packard Instruments, Downers Grove, IL) . Nonspecific binding (i.e. incubation mixture without receptor) was subtracted from all sample values.

For comparison, control tests were also run on samples using prior art techniques. In this regard, a modified (M. Dame et al^, 82 P.N.A.S. USA 7825-29 (1985) hydroxylapatite binding assay (D. Williams et al. , 13 Biochemistry 5537-42 (1974); W. Wecksler et al. , 92 Anal. Biochem. 314-23 (1979)) was used to determine the 1,25- (OH)2-[26,27- ³ H]D3 binding activity in samples. Pellets

were counted in BioSafe II counting solution (RPI, Mount Prospect, IL) in a Prias 400 CL/D liquid scintillation counter (Packard Instruments, Downers Gove, IL) .

In an especially preferred form, the assay is run over a two day period as follows: DAY I

1. Start by adding the biotinylated antibody to the minitubes with a repeat pipetter (25 μL/tube) .

2. Add iodinated antibody to each tube with a repeat pipetter (25 μL/tube).

3. Add receptor standard (or samples) to yield a total volume of 150 μL.

4. Mix tubes and cover with parafilm.

5. Incubate overnight at +4°C, preferably on ice. DAY II

6. Add avidin-Sepharose 100 μL of a 50% slurry.

7. Incubate on ice for 1 hour. Mix tubes every 15-20 minutes. _*

8. Wash pellets 3 times with PBS - 0.5% Triton X- 100 by centrifugation at 800 x g for 5 minutes.

9. The radioactivity of the pellets is counted in gamma-cou3iter.

Results

Purified l,25-(OH)2D3 receptor (3) from pig intesti- nal mucoaa was used as a receptor standard to create a standard curve (e.g. "X" level of radioactivity equals

"Y" level of receptor).

The biotin linked antibody and the iodinated anti¬ body are incubated simultaneously with increasing concentrations of receptor standard or sample followed by precipitation with avidin-Sepharose. After testing dif¬ ferent receptor concentrations, a standard curve range of 0-25 fmol was found to give maximum sensitivity without excessive use of reagents. To determine the amount of iodinated antibody to be added to each assay tube, stan- dard curves were obtained where radioactivity added ranged from 10,000 cpm to 100,000 cpm per tube. One hundred thousand cpm was found to be saturating. Since non-saturating amounts of iodinated antibody produced an excellent linear standard curve, routine use of 40,000 cpm was found to give low background (<10%) and high sensitivity. Concentrations of biotinylated antibody and avidin-Sepharose were carefully titrated and used at saturation. Usually 100 μL of a 50% slurry per tube was enough. The two antibodies used could be inter- changed without altering assay results.

Since vitamin D hormone concentration (and thus receptor occupancy with vitamin D) varies in different tissues and preparations, the possibility that occupancy might affect receptor measurement was tested. Concentra- tions of receptor in samples of PNE were determined in the presence or absence of 4 nM l,25-(OH)2D3. Data obtained show no significant difference between samples

with or without hormone.

The present test finds greatest utility in samples other than PNE. For example, intestinal cytosol from vitamin D-deficient rats and normal humans was prepared as described above and receptor content measured by the present assay and by the prior art ligand binding assay. Vitamin D-deficient rat intestinal extracts and human intestinal extracts, respectively, were found to have 2- fold and 5-fold higher receptor content by the present test than by ligand binding assay. In addition, vari¬ ability was markedly lower with the immunoradiometric assay. We have also successfully used the test to measure receptor levels in rat kidney, chick intestine, chorioallantoic membrane from quail, a pig kidney cell line (LLC-PKi , a human promyelocytic leukemia cell line (HL-60) and human breast cancers.

On rare occasion, the present test has given values somewhat lower than by ligand binding methods. We believe this is because of the presence of plasma vita- min D transport protein in the sample that produces a falsely high value by the ligand binding assay. This illustrates another disadvantage of ligand binding assay and advantage of the present invention.

It is believed that the present invention can provide important new information relating to the 1,25- (OH)2D3 receptor. For example, the receptor has been reported to be up-regulated and down-regύlated in

response to hormone, and likely regulated by age, gluco- corticoid, and during development. Also, a receptor mediated anti-cancerous effect of vitamin D appears to exist in different human cell lines in vitro. If vitamin D related compounds are, in fact, used to treat malignancy, the present invention may be used to determine which cancers possess receptor in preparation for therapy. In this regard, immunoassays to measure estrogen and progesterone receptors are already widely used to predict the response to hormone treatment and to establish a prognosis in breast cancer patients.

It will be appreciated that the above description merely represents the preferred embodiments of the inven¬ tion. The generic idea is to have two antibodies (1)(7) acting ^' at two different epitopes (4) (5), one labelled (2), and the other anchorable (6) for easy separation. While radioactive labels are preferred, color (e.g. a perioxidase system), fluorescent, and enzyme (e.g. alkaline phosphatase) labels are also possible. The anchored antibody' can also instead be attached to other types of beads, plastics, or particles, albeit the biotin system is highly preferred. For example, one might try adapting a magnetic particle system where the particle is anchored to one antibody (cf. Serono Diagnostics of Randolph, Massachusetts which sells one type of magnetic anchoring system under the name "Prolactin MAIA" where precipitation is caused by a magnet). Further, the

-16-

specific antibodies discussed are merely two of many possible antibodies. Thus, the claims should be looked to to assess the full scope of the invention.

Industrial Applicability

The present invention has industrial applicability to diagnosing diseases, prescribing treatments, and moni¬ toring and selecting drug treatments based on vitamin D variants.